Onto-SPELTRA: A robotic assistant based on ontologies and agglomerative clustering to support speech-language therapy for children with disabilities

© Springer International Publishing AG 2017. According to the United Nations International Children’s Emergency Fund (UNICEF), nowadays in several developing countries there is a lack of essential services for children with disabilities in important fields such as speech-language therapy, physiotherapy, and sign language instruction. Likewise, the UNICEF points that in low income countries only between 5 and 15% of children and adults who require assistive technologies have access to them. In this line, in this paper we present a system that provides three main functionalities for speech-language therapists: a decision support system for planning therapy sessions (ontologies), a robotic assistant to motivate children with disabilities to work in therapy activities, and a module to automatically create groups of patients with similar profiles and needs (agglomerative clustering). In order to validate our proposal we worked with two groups: a first one consisting on 111 youth to validate the robot’s appearance and functionalities, and a second group of 70 children with communication disorders and disabilities to determine their response to therapies provided through robot. In the same way, we have used the children’s profiles to populate the ontology and feed the decision support module based on clustering.